ALPPS for hepatocarcinoma under cirrhosis: a feasible alternative to portal vein embolization

Abstract

Hepatocellular carcinoma (HCC) is one of the most common and malignant tumors. Preoperative portal vein embolization (PVE) is currently the most accepted treatment before major hepatic resection for HCC in patients with liver fibrosis or cirrhosis and associated insufficient future liver remnant (FLR). In the last decade, associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) technique has been described to obtain an increase of volume regarding PVE and a decrease of drop out. The initial excessive morbidity and mortality of this technique have decreased drastically due to a better selection of patients, the learning curve and the use of less aggressive variations of the original technique in the first stage. For both techniques a complete preoperative assessment of the FLR is the most important issue and only patients with and adequate FLR should be resected. ALPPS could be a feasible technique in very selected patients with HCC and cirrhosis. As long as it is performed in an experienced center could be used as a first choice technique versus PVE or could be used as a rescue technique in case of PVE failure.

Introduction

Hepatocellular carcinoma (HCC) is one of the most common and malignant tumors with incidence rate ranking the fifth and mortality rate the third among the malignant tumors in the world (1,2). In cases, due to liver cirrhosis, sequential progression from fibrosis to cirrhosis culminates in HCC due to the preneoplastic setting of the cirrhotic background provides a conductive environment for cellular transformation (3,4).

At present, the treatments for HCC include mainly liver resection, liver transplantation, radiofrequency ablation, transcatheter arterial chemoembolization (TACE), and drug therapy. In case of liver cirrhosis, liver transplantation is the most accepted surgical treatment of HCC, especially in Child B or C stages because it treats both, liver insufficiency and the tumor (5,6). Shortage of grafts is the reason why liver resection is accepted as the first-line of treatment in patients with very early stage (7). Since most patients with HCC have underlying chronic liver disease or cirrhosis, the question of insufficient future liver remnant (FLR) is critical in this cohort to prevent postoperative liver failure (PHLF) (8,9). Most of the cirrhotic patients have an intermediate or advanced stage of the Barcelona Clinic Liver Classification (BCLC), and in cases with well-preserved liver function the only indicated treatments are TACE or Sorafenib (10). Although the results in the literature are controversial, there are some studies with great outcomes in liver cirrhosis and advanced HCC and recently, some authors changed these criteria and claimed that liver resection is the only potential curative treatment (11-13).

Preoperative portal vein embolization (PVE) is nowadays the most accepted treatment before major hepatic resection for HCC in patients with liver fibrosis or cirrhosis and insufficient FLR (14-16). After PVE, the drop-out is approximately 20–30% of patients due to do lack of adequate hypertrophy of the FLR (as a result of either inadequate regeneration capability in those with cirrhosis or the presence of collaterals) (17,18) or tumor progression during the relatively long waiting time after PVE to achieve adequate FLR volume (19,20). In the last decade, associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) technique has been described to obtain an increase of volume regarding PVE and a decrease of drop out (21). One of the major drawbacks is that it has been associated with a high morbidity and mortality (22) but in recent years the improvement in the selection of patients and the appearance of less invasive modifications from the original technique have improved the results (23,24). However, the use of ALPPS to induce FLR hypertrophy in HCC patients with cirrhosis and chronic liver disease continues to present some uncertainty. We reviewed the literature about ALPPS in HCC in cirrhosis focusing on volumetric date, drop out and postoperative outcomes and then, and we compared this data with the outcomes of PVE in those patients according to the most relevant studies published in this field.

Preoperative assessment of the future liver remnant

Subsequent early retrospective studies have shown that preoperative assessment of FLR is in fact able to predict hepatic dysfunction in patients undergoing major liver resection (25-27). Considering the liver volume is only a surrogate for liver function and given that liver function may not be homogenous throughout the liver and can be compromised in case of parenchymal liver disease, quantitative functional assessment using regional liver function tests like mebrofenin and albumin scintigraphy has been proposed as a complimentary and possibly a superior methodological approach to prediction of PHLF (28,29). An alternative to measured FLRV or estimated FLRV is the liver remnant volume to body weight ratio (FRLV-BWR), in which the remnant liver weight is calculated as percentage of body weight. A ratio below 0.5% (healthy livers) and 0.7 or 1.1 (cirrhotic livers) has been shown to correlate with adverse outcomes (30).

PVE

From that first series (31), the indications have been significantly extended to make PVE the “gold standard” for patients with large unilobar tumors, with insufficient FLR, and requiring a major hepatectomy. It is a safe procedure and the few alterations produced are probably due to the fact that the hepatocytes in the hepatic “deportalized” lobe experience a process of apoptosis instead of cell necrosis (32,33). On the other hand, the proliferative stimulation represented by the redirection of the portal flow towards the contralateral hepatic lobe induces a growth factor-mediated hyperplasia. The most potent is the hepatocyte growth factor (HGF) that, together with other mediators, stimulates the production of cytokines such as interleukin 6 and TNFα, triggering a gene response that activates the hepatocyte cell cycle and consequently its regeneration (16).

PVE and cirrhosis

The indication criteria for preoperative PVE for HCC under cirrhosis described by Azoulay et al. in 2000 (34) were: patients younger than 70 years, albumin ≥3 g/dL, total bilirubin <2 mg/dL, Quick ≥80%, Indocyanine Green Retention (ICGR) less than 10 and a FLR lower than 40%. Clinically, the increase of the FLRV in cirrhotic livers after PVE is reported to be in the range of 25% to 30%, and the hypertrophy ratio of the FLR has also been reported to be approximately 1.3 to 1.5 (35,36). In a prospective study, Farges et al. (37) suggest that patients with liver cirrhosis prior to partial hepatectomy could benefit from preoperative PVE and recommended performing PVE in patients with right hepatic HCC as a routine preoperative preparation. Recently, Sun et al. (38) compared cirrhotic with non-cirrhotic with PVE and no significant difference was identified between the two groups with respect to FLR volume enlargement at 4–6 weeks following PVE. In a systematic review, the authors (39-42) evaluated the effect of cirrhosis/fibrosis on the hypertrophy response and they described an increase of volume ranging between 24.4–38.4% in cirrhosis versus 39.4–49.6% in non-cirrhosis at a median of 36 days. In cirrhotic patients, the drop-out was 20% with a resectability of 96% (Table 1).

Table 1. Review of the most important articles on portal vein embolization in hepatocellular carcinoma in cirrhotic patients.

Author	Year	n	Histology	Etiology	Gender	Age	Major morbidity (%)	PHLF (%)	Mortality (%)	Stage liver disease	Volumetric data	Drop out (%)
Azoulay et al. (34)	2000	10	7 cirrhosis	10 HCC	10 M	61±11	30	0	0	–	40%±23%	10
Farges et al. (37)	2003	14	7 cirrhosis; 7 chronic disease	14 HCC	12 M: 2 F	60±10	–	1	7.1	–	35%±28% (range, 18–68%) in 4-6 weeks	14.2
Ko et al. (40)	2003	22	11 cirrhosis	HCC and non HCC	–	–	–	9	4.5	–	38.1% in 2 weeks	–
Denys et al. (41)	2005	40	11 cirrhosis	15 alcohol; HBV 5; HBC 15; others 5	31 M: 9 F	62 [35–74]	–	–	–	40 Child A	41%±32% in 4–5 weeks	10
Cotroneo et al. (42)	2009	25	7 cirrhosis	7 HCC	–	–	–	–	–	6 Child A and 1 Child B	32.1% in 4 weeks	8
Sun et al. (38)	2018	21	12 cirrhosis	10 HCC (HBV); 2 IHCC	20 M: 1 F	54.3±8.8	–	–	–	12 child A	31.1%±16.1% increase in 4–6 weeks	–

N, number of patients; PHLF, postoperative hepatic liver failure; HCC, hepatocellular carcinoma; HBV, hepatitis B virus; HVC, hepatitis C virus; IHCC, intrahepatic cholangiocarcinoma; M, male; F, female.

After PVE liver regeneration may be impaired, causing concerns about tumor progression, this is why some authors speculate to perform a preoperative selective TACE in a standardized sequential manner (43,44). Ogata et al. (45) demonstrated that TACE combined with PVE might effectively induce hypertrophy of FLRs in patients with chronic liver disease and improve the 3-year disease-free survival rate. In summary, the main advantage of the PVE is achieving a hypertrophy that ranges between 20–50% in most series, which allows reaching a resectability of 70–100%. However, the two main drawbacks are the risk of tumor progression during the long time of hypertrophy and the absence of liver regeneration. After observing that in most series there are patients with tumor growth, both in the embolized lobe and in the liver.

ALPPS

The higher and more rapid regeneration of ALPPS than PVE could be related to the occlusion of the intrahepatic circulation, which directs the portal flow to the FLR, associated with venous ischemia of the right lobe. The first publication of the World Registry reported an impressive 80% increase of volume (49–116%) in a 7-day interval, with a resectability of 98%, with a total mortality of 9% and 2% died after the first stage (46). These results have improved in the last two years due to several factors: refinements in strategic details such as parenchyma splitting (Tourniquet, microwave, partial-ALPPS, mini-ALPPS, laparoscopic), extension of interval to 15 days, testing the in FLR function interval. But, the oncological outcomes remain unclear (24). Recently, dual embolization has been described, performing firstly a right PVE and after a right hepatic vein embolization but the experience with this technique is still scarce, especially in cirrhotic patients (47).

Short terms outcomes of ALPPS and HCC in cirrhosis

The use of the ALPPS technique in colorectal liver metastases is the most frequently accepted indication (48,49), being more controversial in primary hepatic tumors, especially in intrahepatic cholangiocarcinoma. D’Haese et al. (50) reported higher 90-day mortality among HCC patients than among those with CRLM (31% vs. 7%) and they concluded that the risk associated with ALPPS remains excessive for most HCC patients and that ALPPS should only be performed in a highly selected group of HCC patients younger than 60 years and with a low fibrosis grade. Vennarecci et al. (51) shown ALPPS that it was feasible and safe in HCC patients with cirrhosis and a significant volume increment of FLR can be induced in a short time to allow for completion of the two stage strategy. The same group, in 2016 (52) compared ALPPS in liver cirrhosis (8 HCC) and non-liver cirrhosis (3 CRLM and 1 cholangiocarcinoma). They concluded that ALPPS induced a considerable and comparable FLR growth in HCC patients with liver cirrhosis and patients with CRLM and cholangiocarcinoma with normal liver parenchyma, but the sample size was very small. Chan et al. (53) described in 2016 a safety ALPPS procedure by anterior approach in 17 patients with hepatitis B-related HCC. All patients proceeded to second-stage hepatectomy, after a median of 6 days and a FLR increase of 38.5%. The major morbidity was 11.8% and hospital mortality rate was 5.9% and liver histology confirmed chronic hepatitis in 8 patients and cirrhosis in the remaining 9 patients.

Serenari et al. (54) reported a volume increase of 50% within a median of 7.5 days in 6 cirrhotic livers but one of these 6 patients died within 90 days of ALPPS. In this study, livers from patients who received preoperative chemotherapy (56%, range: 15–227%) or had cirrhosis (50%, range: 14–178%) developed a lower degree of hypertrophy. The authors conclude that, although, there is considerable risk in cirrhotic patients, ALPPS may be a salvage option in selected patients with HCC and fibrosis or cirrhosis and in whom PVE has been unsuccessful. On the other hand, Chan et al. (55) compared complete and partial split in ALPPS and they demonstrate that complete split induced a more significant FLR growth than partial split in chronic hepatitis in terms of daily hypertrophy rate and gain in FLR ratio. The authors also showed that ALPPS could induce FLR hypertrophy in cirrhotic livers within a short period of time. However, even though complete split tended to induce a more rapid FLR hypertrophy than partial split in cirrhosis (hypertrophy rate 32.2 vs. 16.9 mL/day) the difference was less obvious for cirrhotic livers (FLR% increment: 14.8% vs. 11.0%) than for chronic hepatitis (FLR% increment: 18.1% vs. 11.3%).

In terms of ALPPS for HBV-related HCC the results are controversial. Cai et al. (56) described worst outcomes for HBV-related HCC with cirrhosis. They showed that FLRs of cirrhotic liver do hypertrophy with an increase of 28.1%, but it took longer for the FLR to reach to a safe volume to undergo the second operation. Half of the patients died (two of them after the first stage) because of postoperative liver failure or other complications, which emphasizes the need of optimal patient selection to reduce the mortality. The subgroup analysis merely demonstrated that too small FLR (<30%) before operation contributed to a tragic outcome, while the FLR/SLV between 30% and 40% presented a satisfactory result. On the other hand, Wang et al. (33) analyzed the ALPPS outcomes in 45 patients with unresectable hepatitis B virus-related HCC with better results. The majority of patients presented a BCLC A (42.2%) and the grades of liver fibrosis were absence (4.4%), low (22.2%), moderate (24.4%), severe (11.1%) and cirrhosis (28.9%). They concluded that ALPPS could induce enough volume of the FLR to increase to allow liver resection in HCC patients and that the rate of hypertrophy of the FLR negatively correlated with the severity of fibrosis/cirrhosis. In this study, the hypertrophy of FLR in the normal liver was attributed to both regeneration and increased size of hepatocytes, whereas hypertrophy of FLR in the advanced fibrosis liver mainly relies on increase of the size of the hepatocyte (Table 2).

Table 2. Series on the revised articles on ALPPS in hepatocellular carcinoma cirrhosis that included histological features.

Author	Year	n	Histology	Etiology	Gender	Age	Major morbidity (%)	PHLF (%)	Mortality (%)	Stage liver disease	Volumetric data	Drop out (%)
Vennarecci et al. (52)	2016	8	8 cirrhosis	6 HCV; 2 HBV	8 M	65 (36-74)	20	12.5	12.5	8 child A and MELD 8 [7–9]	71.1% increase in 8 (7–10 days)	0
Chan et al. (53)	2016	17	9 cirrhosis; 8 chronic disease	17 HBV	16 M: 1 F	62 [50–80]	11.8	0	5.9	–	38.1% increase in 6 (5–22days)	0
Serenari et al. (54)	2016	8	6 cirrhosis	–	6 M: 2 F	56 [36–74]	12.5	37.5	60	6 Child A and MELD 7 (6-10)	50% increase [14–178%]	0
Chan et al. (55)	2017	25	13 cirrhosis; 8 chronic disease	1 HCV; 21 HBV; 3 Steatohepatitis	23 M: 2 F	62 [50–80]	16	4	8	–	18.1% and 11.3% increase in chronic disease; 14 and 11% increase in cirrhosis	0
Wang et al. (57)	2017	10	9 cirrhosis; 1 chronic disease	10 HBV	9 M: 1 F	41 [33–60]	20	0	10	10 Child A and MELD 7 [0–19]	47% (40–58%) in 28 days (range, 13–31 days)	20
Cai et al. (56)	2017	12	12 cirrhosis	10 HBV	10 M: 2 F	43 [32–79]	58.3	50	50	10 Child A, 1 Child B and MELD 8 [7–13]	38.5% (10.1% to 74.6%) in 10.5 [7–44] days	16.6
Wang et al. (33)	2018	45	13 cirrhosis	45 HBV	40 M: 5 F	52 [24–67]	15	15.5	11.1	Median Child A: 5 [5–6]	Median KGR was 9.6%, in 14 days (range, 7–28 days)	8.9

PHLF, postoperative hepatic liver failure; HBV, hepatitis B virus; HVC, hepatitis C virus; KGR, kinetic growth rate; M, male; F, female.

Due to the scarce literature on this topic, in series of published clinical cases that have described the results of ALPPS in HCC, we have analyzed those that describe the histopathological findings (58-64). A total of 7 cases were registered (5 of cirrhosis and 2 of moderate fibrosis). There was no mortality and only one patient presented small for size, probably because the FLR prior to the second time was 24%. With the exception of this case, the rest of the patients reached an FLR between 29% and 61% in a time range between 4 and 20 days (Table 3).

Table 3. Clinical cases on the revised articles on ALPPS in hepatocellular cirrhosis that included histological features.

Author	Year	Histology	Etiology	Gender	Age	Morbidity	PHLF	Mortality	FLR
Cavaness et al. (64)	2013	Moderate fibrosis	HCV	F	57	No	No	No	17% to 33% in 4 days
Chia et al. (58)	2014	Moderate fibrosis	HBV	M	55	No	No	No	26.8% to 37.4% in 8 days
Xiao et al. (59)	2015	Cirrhosis	–	–	–	No	No	No	27% to 40.6% in 13 days
Cheung et al. (60)	2016	Cirrhosis	HBV	M	55	No	No	No	22% to 29% in 7 days
de Santibañes et al. (61)	2016	Cirrhosis	–	F	66	No	No	No	40% to 61% in 10 days
Papamichail et al. (62)	2016	Cirrhosis	Alcohol	M	68	No	Small for size	No	14% to 24% in 10 days
Chen et al. (63)	2016	Cirrhosis	HBV	M	43	No	No	No	29.1% to 51.2% in 20 days

PHLF, postoperative hepatic liver failure; HBV, hepatitis B virus; HVC, hepatitis C virus; FLR, future liver remnant; M, male; F, female.

Long terms outcomes of ALPPS and HCC in cirrhosis

Few authors detail the follow-up results and the patients included in the follow-up are scarce. Senerari et al. (54) describe an overall survival (OS) of 75% and a disease free survival of 62% at 1 year. Wang et al. (33) published an OS rate at 1- and 3-year of 64.2% and 61.2% whereas the DFS rates at 1- and 3-year were 64.2% and 61.2%, respectively. On the study by Cai et al. (56) at a mean follow-up of 16 months, 1 patient died of upper gastrointestinal hemorrhage at 4 month and another patient died of recurrence and lung metastasis at the 13 months, and 4 out 6 were still alive and 2 were live and free of disease. In the series (n=10) of Wang et al. (57), the 2 patients who did not complete de stage 2 died in the following 3 months after discharge. Another two cases had a tumor recurrence within 3 months of discharge and 1 of these died and another underwent TACE with Sorafenib. Another patient subsequently underwent TACE when local tumor recurrence was observed at 9 months and 4 cases had no disease recurrence.

Role of ALPPS variants modifications in HCC

The variants of the original ALPPS technique have allowed to improve the results and this improvement in the morbidity it could be useful in high risk patients such as cirrhotic patients. All the variants share the same theoretical concept from the original technique. The only novel aspect that these variants introduce concerns the anatomical place of the bipartition and, therefore, the type of hepatectomy that will be carried out during the second surgical time. These alternatives are based on two truly novel and different concepts with respect to classical ALPPS and consist of not splitting the hepatic parenchyma (ALPPS-Tourniquet, radiofrequency ablation, microwave ablation) or splitting the parenchyma partially (partial ALPPS, Mini ALPPS). The first variant described was our technique [2011] named as Tourniquet-ALPPS or ALTPS. We have operated on 4 HCC treated with Tourniquet-ALPPS (2 with cirrhosis and 2 with low fibrosis). Despite being a very small sample, both groups presented a similar FLR (42.5% and 105% vs. 51.7% and 187.5%, respectively) in the same period without mortality, with a major complication in each group and without drop out. Wang et al. (57) described RALPPS technique in cirrhosis-related HCC from HBV. Nine of 10 patients included in the study present with cirrhosis. Two patients did not proceed to the second-stage operation: one patient had liver dysfunction and massive ascites after stage I, and the other patient had metastasis in FLR tissue during the waiting period before stage II. The median FLR before stage I was 31% (364 mL) with an increased to 47% (632 mL) before stage II after a median interval of 28 days and a median percentage increase in FLR of 53% (210 mL). The morbidity and mortality results were poor with a PHLF in 5 of 10 patients and mortality in 5 of 10 patients. The results were related due to the patients were operated on in the II stage with and insufficient FLR volume for a liver cirrhosis.

ALPPS and histology features

Regarding liver histology before the first intervention, it is considered that patients with fibrosis have worse regeneration and that it is specially related to the degree of fibrosis. There are discrepancies regarding the regeneration capacity of the liver when it presents with fibrosis, cirrhosis, cholestasis, macrosteatosis and alterations related to chemotherapy (Figure 1). Furthermore, there are no data regarding the influence of histological alterations before the second intervention and the presence of postoperative hepatic insufficiency. Several studies have been able to establish a histological correlation with the increase of FLR in patients with ALPPS by using immunohistochemical techniques such as Ki-67 for the measurement of hepatocyte cell proliferation index (65-69). The underlying biological substrate is not well understood, but it is suggested that a combination of factors including hemodynamic changes or damage caused by the surgery itself would be able to induce a systemic response mediated by growth factors, transcription and cytokines that would ultimately lead to the activation and proliferation of hepatocytes (66,70). Histologically, differences have been observed with respect to PVE (69,70), as an increase in the hepatocyte density or a decrease in the size of the hepatocyte, as well as a greater cellular atrophy, degeneration, necrosis, fibrosis or sinusoidal dilatation in the deportalized areas, as well as congestion. The latter would be of special relevance to explain the higher proliferation rates reported in ALPPS. However, little has been studied about ALPPS in a cirrhotic liver. Cirrhosis is the last phase of a dynamic diffuse fibrosing process in which the normal architecture of the liver is replaced by a nodular pattern as a result of liver damage, usually chronic (71). The fibrotic tissue ends up hindering liver regeneration (72). Although previous cirrhosis would limit in principle the process of hypertrophy, the degree of hypertrophy induced by ALPPS could be beneficial even for cirrhotic patients (59,73,74). However, discordance between the growth in liver volume in ALPPS and the functional growth of the proliferated liver has been detected, with multiple signs of immaturity in the tissue being appreciated (73,75-77). Whether these discrepancies are transient or not, this means that care must be taken in patients with a low previous liver remnant, as well as in cirrhotic patients. It is unknown if this discordance persists also in modified ALPPS techniques, since there are practically no studies that address liver function for this technique (76). To solve these unknowns, different animal models are being developed for the study and understanding of this set of processes (78-81), included a rat model with fibrosis.

Figure 1.

Biopsies in the future liver remnant in a liver without alterations, before (A) and after (B) Tourniquet ALPPS (H&E, 100×). The normal architecture of the liver is preserved in both and there are no significant differences between them. (C) Second stage biopsy of cirrhotic liver (Masson’s Trichrome, 40×) after Tourniquet ALPPS. In contrast to the previous images, the normal architecture of the liver is replaced by multiple regeneration nodes separated from each other by fibrous septa, and no relevant differences can be highlighted when compared with a cirrhotic liver previous to the surgery (D, H&E 40×).

Future directions

Major liver resections in cirrhotic livers continue to be complex interventions due to the high risk of PHLF and mortality. The most important and key aspect to obtain good results in this subpopulation of patients is to achieve a hepatic remnant with enough size to perform the surgery with the greatest possible safety. The study of the patient should be as complete as possible, therefore it is recommended that along with the volumetric and the ICGR15 rate, the RLV-BWR and the quantitative functional assessment using regional liver function tests like mebrofenin and albumin scintigraphy should also be performed. The patients in which all these measurements are favorable, surgery could be performed with the lowest probability of failure. Regarding the type of approach in these patients with a cirrhotic liver, the PVE has traditionally been associated with or not associated with TACE as a step prior to surgery, but the appearance of the ALPPS technique offers a new alternative. The great advantage of this technique is that it provides us with high rates of resectability due to its large volumetric capacity in a short time compared to the PVE. The initial problems of this technique on the morbidity and mortality results have decreased drastically due to a better selection of patients, learning curve and less aggressive techniques variations in the first time that provide the same results of volumetric increase. For this reason, ALPPS is a feasible technique in patients selected with HCC and cirrhosis, and it could be performed in an experienced center as a technique of first option in cases with a very low FLR or as a rescue technique in the face of PVE failure.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.